A few days ago, the Eighth Circuit became the first appellate court to interpret Halliburton Co. v. Erica P. John Fund, Inc., 134 S. Ct. 2398 (2014) (Halliburton II), relying on that case to reverse a district court’s class certification order in IBEW Local 98 Pension Fund v. Best Buy Co.
The case is interesting because it has an unusually clean fact pattern for analyzing some basic dilemmas in the law governing Section 10(b) actions.
The facts are these:
On the morning of September 14, 2010, before the market opened, Best Buy issued a press release that increased its full-year earnings guidance. By the time of the opening, its stock price was up 7.5% from the prior day’s close. Two hours after the press release issued, Best Buy held a conference call with market analysts, during which the CFO stated that the company was “on track to deliver and exceed” its EPS guidance.
On December 14, 2010, Best Buy issued a press release announcing a decline in sales and reduction in EPS guidance, causing a stock price decline.
The plaintiffs alleged that both the September 14 press release, and the subsequent conference call, were fraudulent. The district court held that the press release was immunized as a forward-looking statement, accompanied by sufficient cautionary language, under the PSLRA’s safe harbor. However, the court held that the “on track” representation was not forward looking, and claims based on that statement could proceed.
The problem, however, as the plaintiffs’ own expert eventually opined, was that Best Buy’s stock price increased after the immunized press release, and did not appear to react to the earnings conference call. The plaintiffs’ expert also opined that the conference call conveyed information that was “virtually the same” as the information in the press release.
The plaintiffs offered two theories to explain how the conference call may have impacted Best Buy’s stock price. First, they claimed that the conference call caused an upward earnings drift over the next several weeks. And second, they claimed that the “on track” confirmatory statements served to maintain Best Buy’s stock price, already boosted by the press release. Accepting this evidence, the district court certified the class.
On appeal, the Eighth Circuit reversed. It concluded that, in accordance with Halliburton II, the defendants had rebutted the presumption that the conference call impacted Best Buy’s price. In the Eighth Circuit’s view, because the plaintiffs’ own expert agreed that the stock price had only increased in response to the immunized press release, and agreed the conference call conveyed no new information, the conference call could not have had an effect. The court rejected the “earnings drift” theory as contrary to the efficient market hypothesis, but did not – in explicit terms – weigh in on the plaintiffs’ price maintenance theory, except to say that this was unlike situations where a third-party confirms an earlier corporate statement.
Judge Murphy, in dissent, faulted the majority for failing to directly confront the plaintiffs’ price maintenance theory – which, she believed, had not been rebutted.
There are several interesting things to comment on here.
[More under the jump]
First, notice that despite being immunized by the safe harbor, the plaintiffs’ evidence showed – and the court apparently agreed – that the earnings projection in the press release impacted Best Buy’s stock price. This demonstrates the power of so-called “cautionary language” to protect defendants from liability for fraud. The old “bespeaks caution” doctrine on which the safe harbor was modeled only allowed cautionary language to immunize fraudulent projections when the language was sufficient to render the projections immaterial. See Shaw v. Digital Equip. Corp., 82 F.3d 1194 (1st Cir. 1996). But as the district court interpreted the safe harbor in this case, even a statement that was intentionally designed to mislead investors, and was successful in accomplishing that goal, became untouchable. This is exactly why the First Circuit described the safe harbor as granting “(within limits) a license to defraud” and cautioned that the safe harbor must be given a “restricted” interpretation. In re Stone & Webster, Inc., Sec. Litig., 414 F.3d 187 (1st Cir. 2005).
Second, the case provides a unique set of facts because the immunized projections were issued only two hours before the reaffirming statements. Price maintenance theory usually proceeds on the idea that after an initial false statement is made, time passes, new information enters the market, and the older, stale statement naturally begins to exert less of a pull on market prices. By interjecting confirmatory statements into the mix of information, the company revives the old information, reaffirms its continued validity, and thereby keeps its stock price inflated. A useful demonstration of the concept can be found in the case of Hunt v. Enzo Biochem, Inc., 530 F. Supp. 2d 580 (S.D.N.Y. 2008). There, a company boosted its stock price by fraudulently announcing a plan to open new clinics for a revolutionary HIV treatment. When no additional announcements were forthcoming to confirm the progress of the clinics, the stock price dropped. The court accepted the plaintiffs’ theory that the company’s silence revealed the truth for loss causation purposes: had the company actually made progress on the clinics, it would have said something – a fact that the market recognized. Silence when speech was expected caused a price drop; confirmatory speech would have maintained the price.
Bringing this idea back to Best Buy, things might have looked very different if the press release and the conference call were separated by a few weeks. In that instance, the later statements would be taken not only as a commentary on Best Buy’s status when the projection first issued, but also as a commentary on sales in the intervening period. But with the two statements in such close proximity, there isn’t the same inference that the conference call added anything to the mix.
That said, when examining the likely effect of the conference call statements, I think the critical question should be what would have happened if the defendants had remained silent, as in Hunt. Having issued the projections just hours earlier, if they’d refused to confirm them on the call – if they’d refused to confirm that they were “on track” to meet those projections – the silence may have functioned as a revelation of the truth.
Unfortunately, this thought experiment – what might have been the effect of silence (at least in situations where silence is an option – not always true when affirmative disclosure obligations attach) – is not one that courts generally undertake. For example, in ECA & Local 134 IBEW Joint Pension Trust of Chi. v. JP Morgan Chase Co., 553 F.3d 187 (2d Cir. 2009), the Second Circuit found that JP Morgan Chase’s claims to have “highly disciplined” risk management processes were mere puffery, because “almost every investment bank makes these statements.” In so holding, the court failed to consider what would have happened if JP Morgan Chase had failed to attest to the “discipline” of its risk management. Had the court done so, it might have concluded that silence in the face of an industry standard would have sent a powerful market signal. This is not a new idea – the Supreme Court itself recognized in Basic Inc. v. Levinson, 485 US 224 (1988), that in some contexts, a “no comment” – silence – may function as an admission.
But that line of thinking leads to the final issue – how might the case have looked if the two statements, rather than being made hours apart, had instead been made simultaneously? Courts usually hold that when statements contain forward-looking and historical components, only the forward-looking portion may be immunized by the safe harbor. But that implies that plaintiffs can demonstrate the independent materiality of the historical information. In cases with statements similar to Best Buy’s, some courts have held that an earnings projection by its nature implies that the company is, at that moment, positioned to meet it; further statements along those lines add nothing, and cannot be analyzed separately. Unsurprisingly, then, in Best Buy, the Eighth Circuit expressed doubts that the “on track” language had any independent materiality, going out of its way to note that it had not been tasked with deciding whether the present-tense statements could “meaningfully be distinguished” from the forward-looking ones.
But that folds us into what Amgen Inc. v. Conn. Ret. Plans & Trust Funds, 133 S. Ct. 1184 (2013) supposedly forbids – a materiality determination at the class certification stage. The Eighth Circuit claimed to have been following Amgen, but it’s hard not to suspect that its interpretation of the price impact evidence was dependent upon its perception of the conference call’s materiality. After all, why else would the court have highlighted testimony that the call offered “virtually the same” information as in the press release?
Thus, the real significance of Best Buy is to illuminate the interdependence of the price impact and materiality inquiries – and to demonstrate how far down the rabbit hole Halliburton II leads.